Using the motor drive as a sensor to extract spatially dependent information during servo operation

This paper presents a methodology for extracting useful spatial signals using the motor drive as the sensor during servo operation. Spatially dependent phenomena yield non-stationary frequency content during servo operation. With variable frequency and frequency dependent amplitude in the current and torque signals, the underlying, valuable spatial information is difficult to extract using conventional frequency domain approaches. The methodology presented uses the motor drive to estimate the spatially dependent disturbance torque information during servo operation using spatial domain re-sampling of temporal domain signals internal to the motor drive. Spatial domain information is speed invariant, making it accessible even during servo operation. Additionally, this paper develops a framework to operate through varying speeds to help isolate the systematic, spatially dependent content using spatial synchronous averaging. These methods and concepts are evaluated through simulation as well as experimentation. This methodology can be utilized to provide robust control of machines through identification of spatial components, as well as improve diagnostics of machine and load degradation.